JP4309133B2 - Zinc finger binding domain for nucleotide sequence ANN - Google Patents

Abstract

Polypeptides that contain from 2 to 12 zinc finger-nucleotide binding regions that bind to nucleotide sequences of the formula (ANN)2-12 are provided. Polynucleotides that encode such polypeptides and methods of regulating gene expression with such polypeptides and polynucleotides are also provided.

Description

  The field of this invention is zinc finger proteins that bind to target nucleotides. More particularly, the present invention relates to amino acid residue sequences within the α-helical domain of zinc fingers that specifically bind to a target nucleotide of formula 5 ′-(ANN) -3 ′.

  The construction of artificial transcription factors has been the subject of great interest over the past several years. Gene expression can be specifically regulated by a polydactyl zinc finger protein fused to a regulatory domain.

The zinc finger domains of the Cys ₂ -His ₂ family have so far been most promising for the construction of artificial transcription factors because of their modular structure. Each domain consists of approximately 30 amino acids and folds into a ββα structure stabilized by chelating zinc ions by hydrophobic interactions and conserved Cys ₂ -His ₂ residues. To date, the most clearly characterized protein in this zinc finger protein family is the mouse transcription factor Zif268 [Pavletich et al., (1991) Science 252 (5007), 809-817, Elrod- Erickson et al., (1996) Structure 4 (10), 1171-1180]. The analysis of the Zif268 / DNA complex shows that DNA binding is achieved primarily by the interaction of the 3 ', middle and 5' nucleotides of the 3bp DNA subsite with the α-helix amino acid residues at positions 1, 3 and 6, respectively. Suggests that Positions 1, 2 and 5 have been shown to have direct or water-mediated contact with the phosphate backbone of DNA. Leucine is usually found at position 4 and is packaged into the hydrophobic core of the domain. Position 2 of the α helix has been shown to interact with other helix residues and may also have contact to nucleotides outside the 3 bp subsite [Pavletich et al., (1991) Science 252 (5007). 809-817, Elrod-Erickson et al., (1996) Structure 4 (10), 1171-1180, Isalan, M. et al., (1997) Proc Natl Acad Sci USA 94 (11), 5617-5621].

  The selection of modular zinc finger domains that recognize each of the 5′-GNN-3 ′ DNA subsites with high specificity and affinity and their sophistication by site-directed mutagenesis have been demonstrated to date. These modular domains can be assembled into zinc finger proteins that recognize extended 18 bp DNA sequences that are unique within the human or any other genome. Furthermore, these protein functions function as transcription factors, can alter gene expression when fused to the regulatory domain, and can even be made hormone-dependent by fusion to the ligand binding domain of the nuclear hormone receptor. . It is important to extend the existing set of modular zinc finger domains to recognize each of the 64 possible DNA triplets in order to allow rapid construction of zinc finger-based transcription factors that bind to DNA sequences . This goal can be achieved by phage display selection and / or rational design.

Due to the limited structural data on zinc finger / DNA interactions, rational design of zinc proteins is a waste of time and in many cases may not be possible. Furthermore, the most naturally occurring zinc finger protein is composed of a domain that recognizes the 5′-GNN-3 ′ form of the DNA sequence. Only a few zinc finger domains are found in the native protein that bind to sequences of the 5′-ANN-3 ′ type, eg Gfi-1 finger 5 (5′-AAA-3 ′) [Zweidler-McKay et al., (1996) Mol. Cell. Biol. 16 (8), 4024-4034], YY1 finger 3 (5'-AAT-3 ') [Hyde-DeRuyscher, et al., (1995) Nucleic Acids Res. 23 ( 21), 4457-4465], fingers 2 and 6 of CF2II (5 ′-[A / G] TA-3 ′) [Gogos et al., (1996) PNAS 93, 2159-2164] and finger 2 of TTK ( 5'-AAG-3 ') [Fairall et al., (1993) Nature (London) 366 (6454), 483-7]. However, in the structural analysis of the protein / DNA complex by X-ray or NMR test, no interaction between the amino acid residue at position 6 of the α helix and nucleotides other than 5 ′ guanine was observed. Thus, the most promising identification method for novel zinc finger domains that bind to DNA target sequences of the 5′-ANN-3 ′, 5′-CNN-3 ′ or 5′-TNN-3 ′ type is the phage display It is a selection by. A limiting step for this method is the construction of a library that allows the specification of 5 ′ adenine, cytosine or thymine. Phage display selection was based on Zif268 where the various fingers of this protein were randomized [Choo et al., (1994) Proc. Natl. Acad. Sci. USA 91 (23), 11168-72, Rebar et al., (1994) Science (Washington, DC, 1883) 263 (5147), 671-3, Jamieson et al., (1994) Biochemistry 33, 5687-5695, Wu et al., (1995) PNAS 92, 344-348, Jamieson et al., (1996) Proc Natl Acad Sci USA 93, 12834-12839, Greisman et al., (1997) Science 275 (5300), 657-661]. A set of 16 domains that recognize 5′-GNN-3 ′ type DNA sequences is C7 finger 2, a derivative of Zif268 [US Pat. No. 6140081, the disclosure of which is incorporated by reference, Wu et al., (1995 ) PNAS 92, 344-348 Wu, 1995 # 164] has been previously reported from a randomized library [Segal et al., (1999) Proc Natl Acad Sci USA 96 (6), 2758-2863]. . In such a strategy, the selection recognizes 5'-GNN-3 'or 5'-TNN-3' because Asp ^{2 of} finger 3 contacts the 5 'guanine or thymine complementary base at the finger 2 subsite [Pavletich et al., (1991) Science 252 (5007), 809-817, Elrod-Erickson et al., (1996) Structure 4 (10), 1171-1180]. The limited modularity of the zinc finger domain, which can recognize nucleotides outside the 3 bp subsite in some cases, has been well studied [Wolfe et al., (1999) Annu. Rev. Biophys. Biomol. Struct. .3, 183-212, Segal et al., (2000) Curr Opin Chem Biol 4 (1), 34-39, Pabo et al., (2000) J. Mol. Biol. 301, 597-624, Choo et. al., (2000) Curr.opin.Struct.Biol. 10, 411-416]. One way to overcome the limitations imposed by target site overlap is the randomization of amino acid residues in two adjacent fingers [Jamieson et al., (1996) Proc Natl Acad Sci USA 93, 12834-12839, Isalan. et al., (1998) Biochemistry 37 (35), 12026-12033]. A second but time-consuming method is the sequential selection of fingers 1-3 for specific 9 bp target sites that account for the individual structure and DNA binding scheme of each finger and its surrounding fingers [Greisman et al , (1997) Science 275 (5300), 657-661, Wolfe et al., (1999) J Mol Biol 285 (5), 1917-1934].

  Because this method is based on the modular nature of zinc finger domains, it allows for the rapid construction of zinc finger proteins by the academic community, with limited cases of concern about the limitations imposed by cross-subsite interactions. It is proved that it is only a given number. This disclosure introduces a new strategy for the selection of zinc finger domains that specifically recognize 5′-ANN-3 ′ type DNA sequences. The specific DNA binding properties of these domains were evaluated by a multi-target ELISA on all 16 5′-ANN-3 ′ triplets. These domains can be easily incorporated into polydactyl proteins containing various numbers of 5′-ANN-3 ′ domains, each capable of specifically recognizing an extended 18 bp sequence. Furthermore, these domains could specifically alter gene expression when fused to regulatory domains. These results foresee the possibility of constructing polydactyl proteins from pre-specified building blocks. Furthermore, the domains characterized herein greatly increase the number of DNA sequences that can be targeted by artificial transcription factors.

BRIEF SUMMARY OF THE INVENTION The present disclosure teaches the construction of a novel phage display library that allows the selection of zinc finger domains that recognize 5'-ANN-3 'type DNA sequences. Such domains were isolated and showed sensitive binding specificity for the 3 bp target site from which they were selected. These zinc finger domains were implanted into 6-finger proteins that specifically bind to their 18 bp target site with affinities ranging from pM to low nM. When fused to a regulatory domain, an artificial 6-finger protein containing five 5'-ANN-3 'and one 5'-TNN-3' domains is a minimal promoter containing a zinc finger binding site and a TATA box The luciferase reporter gene was regulated under the control of Furthermore, 6-finger proteins assembled from 5'-ANN-3 'and 5'-GNN-3' domains showed specific transcriptional regulation of the endogenous erbB-2 and erbB-3 genes, respectively. These results indicate that modular zinc finger domains that bind to 3 bp target sites other than 5'-GNN-3 'can be selected, and that they are suitable as additional modules to create artificial transcription factors, Shows that the number of sequences that can be targeted by a DNA binding protein constructed from a pre-defined zinc finger domain is greatly increased.

That is, the present invention includes 2 to 12 zinc finger nucleotide binding peptides, wherein at least one of the peptides includes a nucleotide binding region having any one of SEQ ID NOs: 7 to 71 and 107 to 112, and A purified polypeptide is provided. In a preferred embodiment, the polypeptide comprises 2-6 zinc finger-nucleotide binding peptides. Such a polypeptide binds to a nucleotide comprising the sequence 5 ′-(ANN) _n -3 ′, wherein each N is A, C, G or T and n is 2-12. Preferably, each of the peptides binds to a different target nucleotide sequence. The polypeptides of this invention can be operably linked to one or more transcriptional regulators, such as a repressor or activator.

  Also provided are polynucleotides that encode the polypeptides, expression vectors comprising the polynucleotides and cells transformed with the expression vectors.

In a related aspect, the invention provides a method of modulating the expression of a nucleotide sequence comprising the sequence (5'-ANN) _n- 3 ', where n is an integer from 2-12. The method includes exposing the nucleotide sequence to an effective amount of the polypeptide of the invention under conditions in which the polypeptide binds to a nucleotide expression regulatory sequence. That is, the sequence 5 ′-(ANN) _n −3 ′ can be located within the transcription region of the nucleotide sequence, the promoter region of the nucleotide sequence, or the expressed sequence tag. The polypeptide is preferably operably linked to one or more transcriptional regulators.

FIG. 1 schematically illustrates the construction of a zinc finger phage display library. Solid arrows indicate the interaction between the amino acid residues of the zinc finger helices and the nucleotides at their binding sites as determined by Zif268 x-ray crystallography, and the dotted line indicates the proposed interaction. FIG. 2 shows the amino acid sequence of the finger-2 recognition helix from selected clones. For each DNA target site, several single clones were sequenced after the sixth round of screening and amino acid was determined to evaluate the selection. The DNA recognition subsite for finger 2 is shown to the left of each set, followed by the number of each occurrence. The position of the amino acid residue within the α helix is indicated at the top. The boxed arrangement has been tested in detail and represents the best binder for each set. Sequences marked with an asterisk were additional analytical clones. Clone with Ser ^{4 to} Cys ⁴ mutation in ¹ finger 3. Sequence determined after subcloning the zinc finger sequences from the DNA pool after the sixth round of selection into the ^2- modified pMAL-c2 vector. ^* Additional clones analyzed. FIG. 3 (shown in 26 panel: 3a-3z) shows a multiple target specificity assay to test the DNA binding properties of selected domains. At the top of each graph is shown the amino acid sequence of the finger-2 domain of the analyzed 3-finger protein (positions 2-6 for the helix starting point). The solid bars indicate the target oligonucleotides with different finger-2 subsites: AAA, AAC, AAG, AAT, ACA, ACC, ACG, ACT, AGA, AGC, AGT, ATA, ATC, ATG and ATT. Represents a bond. The white bar represents binding to a series of oligonucleotides that differ only in the 5 ′ position of the finger-2 subsite, eg, 5′-AAA-3 ′ subsite (FIG. 3a) binds to AAA, CAA, GAA, or TAA 5 ′ recognition is evaluated for the domain to do. The height of each bar represents the relative affinity of the protein for each target, averaged over two independent experiments and normalized to the highest signal between black or white bars. The vertical line of error represents the deviation from the average. The analyzed protein corresponds to the helix sequence enclosed in the square from FIG. ^* A protein containing a finger-2 domain generated by site-directed mutagenesis. FIG. 4 (shown in 2 panels: A and B) shows the construction and ELISA analysis of a 6-finger protein containing a domain that recognizes the 5′-ANN-3 ′ DNA sequence. A: 6-finger proteins pAart, pE2X, pE3Y and pE3Z were constructed using the Sp1C framework. The amino acid residues at positions -1 to 6 of the α recognition helix are given for each finger used. B: The protein was expressed in E. coli as an MBP fusion protein. Binding specificity was determined using immobilized biotinylated oligonucleotides (E2X, 5′-ACC GGA GAA ACC AGG GGA-3 ′ (SEQ ID NO: 72), E3Y, 5′-ATC GAG GCA AGA GCC ACC-3 ′ (SEQ ID NO: 73 ), E3Z, 5′-GCC GCA GCA GCC ACC AAT-3 ′ (SEQ ID NO: 74), Aart, 5′-ATG-TAG-AGA-AAA-ACC-AGG-3 ′ (SEQ ID NO: 75)) Analysis was performed by measuring the binding activity from the lysate. The test is performed in duplicate and the bar represents the standard deviation. Black bar: pE2X, striped bar: pE3Y, gray bar: pE3Y, white bar: pAart. FIG. 5 (shown in 2 panels: A and B) shows the results of the luciferase reporter assay. HeLa cells were isolated from the zinc finger expression plasmid (pcDNA as a control) and luciferase under the control of a minimal promoter with a TATA-box and a zinc finger binding site (A: 5 × Aart binding site, B: 6 × 2C7 binding site). Co-transfected with a reporter plasmid containing the gene. Luciferase activity in the cell extract was measured 48 hours after transfection. Each bar represents an average value (+/− standard deviation) of duplicate measurement results. Y-axis: 10 ³ divided by the optical unit. X-axis: construct encoding transfected zinc finger protein, control, reporter alone. FIG. 6 (shown in 2 panels: A and B) shows retrovirus-mediated gene targeting. A431 cells were infected with retroviruses encoding pE2X (A) or pE3Y (B) fused to activation domain VP64 or repression domain KRAB, respectively. Three days later, intact cells were stained with ErbB-1 specific mAb EGFR-1, ErbB-2 specific mAb FSP77, or ErbB-3 specific mAb SGP1 in combination with a phycoerythrin labeled secondary antibody. Dotted line: control staining (primary antibody omitted), dash line: specific staining of mock infected cells, dot / dash line: cells expressing zinc finger protein VP64 fusion, bold line: cells expressing zinc finger protein-KRAB fusion.

Detailed Description of the Invention I Zinc Finger Polypeptides The present invention provides isolated and purified polypeptides comprising 2-12 nucleotide binding domain peptides derived from zinc finger proteins. The nucleotide binding domain peptide is derived from the alpha helix portion of the zinc finger protein. Preferred nucleotide binding domain peptides have the amino acid residue sequence of any of SEQ ID NOs: 7-71 or 107-112. Preferably, the peptide has an amino acid residue sequence of any of SEQ ID NOs: 46-70. More preferably, the peptide has an amino acid residue sequence of any of SEQ ID NOs: 10, 11, 17, 19, 21, 23-30, 32, 34-36, 42, 43 or 45. Each of the peptides appears to specifically bind to a nucleotide target sequence corresponding to Formula 5′-ANN-3 ′, where N is any nucleotide (ie, A, C, G or T). Designed and generated. That is, the polypeptide of the present invention binds to the nucleotide sequence 5 ′-(ANN) _n -3 ′ (where n is an integer of 2 to 12). Preferably, n is 2-6.

  The compounds of this invention are isolated zinc finger-nucleotide binding polypeptides that bind to and modulate the function of the ANN nucleotide sequence. A polypeptide can promote or repress transcription of a gene and can bind to DNA or RNA. A zinc finger-nucleotide binding polypeptide refers to a polypeptide that is produced by a mutant form of a zinc finger protein or by genetic engineering. The polypeptide can be, for example, a hybrid comprising zinc finger domain (s) from one protein linked to zinc finger domain (s) of a second protein. The domain can be wild type or a mutation can be introduced. Polypeptides include truncated forms of wild-type zinc finger proteins. Examples of zinc finger proteins from which polypeptides can be produced include TFIIIA and zif268.

  The zinc finger-nucleotide binding polypeptide of the present invention contains a unique heptamer (a sequence of 7 amino acid residues) within the α-helical domain of the polypeptide, which determines the binding specificity of the target nucleotide. To do. The heptamer sequence can be located anywhere within the alpha helix domain, but the residues are conventionally numbered in the art, according to which the heptamer is in positions -1 to 6 It is preferable that it is elongated. The polypeptides of this invention can include β-sheet and framework sequences known in the art to function as part of a zinc finger protein. A number of zinc finger-nucleotide binding polypeptides were produced and tested for binding specificity to target nucleotides including ANN triplets.

  A zinc finger-nucleotide binding polypeptide derivative is derived or produced from a wild-type zinc finger protein by truncation or extension, or by a site-directed mutagenesis method or a combination of these procedures. It can be derived or produced as a variant of The term “truncated” refers to a zinc finger-nucleotide binding protein that contains less than the maximum number of zinc fingers found in a native zinc finger binding protein or lacks unwanted sequences. For example, truncating the zinc finger-nucleotide binding protein TFIIIA, which originally contains nine zinc fingers, can result in a polypeptide with only one to three zinc fingers. Extension refers to a zinc finger polypeptide to which an additional zinc finger module has been added. For example, TFIIIA can be expanded to 12 fingers by adding 3 zinc finger domains. Furthermore, truncated zinc finger-nucleotide binding polypeptides can include zinc finger modules from multiple wild-type polypeptides, and thus can be “hybrid” zinc finger-nucleotide binding polypeptides.

  The term “mutated” refers to a zinc finger derived nucleotide binding polypeptide obtained by performing any of the known methods for achieving random or site-directed mutagenesis of DNA encoding a protein. Say about. For example, in TFIIIA, one or more repeat units of the consensus sequence can be replaced with non-conserved residues by introducing mutations. Mutations can also be introduced into truncated zinc finger-nucleotide binding proteins.

  Examples of known zinc finger-nucleotide binding polypeptides that can inhibit the function of a nucleotide sequence containing a zinc finger-nucleotide binding motif by truncation, extension and / or mutagenesis according to the present invention include TFIIIA and zif268. is there. Other zinc finger-nucleotide binding proteins are well known to those skilled in the art.

The polypeptides of this invention can be produced using various standard techniques known in the art. A zinc finger protein phage display library was generated and selected under conditions that favor the enrichment of sequence-specific proteins. Zinc finger domains that recognize some sequences required refinement by site-directed mutagenesis guided by both phage selection data and structural information. Previously, we identified each of the 5'-GNN-3 'type DNA sequences isolated by phage display selection based on C7, a variant of the mouse transcription factor Zif268 and refined by site-directed mutagenesis A characterization of the 16 zinc finger domain that recognizes itself has been reported [Segal et al., (1999) Proc Natl Acad Sci USA 96 (6), 2758-2863, Dreier et al., (2000) J. Mol. Biol. 303, 489-502]. The molecular interaction of Zif268 and its target DNA 5′-GCG TGG GCG-3 ′ (SEQ ID NO: 76) has been characterized in great detail. In _general, specific DNA recognition of Cys 2 -His ₂ type of zinc finger domains are transmitted by amino acid residues 1, 3 and 6 of the α-helix, all three residues at any case necessarily DNA It is not in contact with the base. One major cross-subsite interaction was observed from position 2 of the recognition helix. Asp ² has been shown to stabilize zinc finger domain binding by direct contact with complementary adenine or cytosine of the 5 ′ thymine or guanine, respectively, of the subsequent 3 bp subsite. These non-modular interactions have been reported as target site overlap. In addition, other interactions between amino acids that create extended binding sites and nucleotides outside the 3 bp subsite have been reported [Pavletich et al., (1991) Science 252 (5007), 809-817, Elrod-Erickson et al. al., (1996) Structure 4 (10), 1171-1180, Isalan et al., (1997) Proc Natl Acad Sci USA 94 (11), 5617-5621].

From the previously reported selection of phage display libraries for zinc finger domains that bind to 5 ′ nucleotides other than guanine or thymine, crossover from aspartic acid at position 2 of the finger-3 recognition helix RSD-E-LKR Satisfactory results were not obtained due to subsite interaction. In order to expand the availability of zinc finger domains for the construction of artificial transcription factors, domains that specifically recognize 5′-ANN-3 ′ type DNA sequences were selected. The other group recognizes four 5′-ANN-3 ′ subsites, 5′-AAA-3 ′, 5′-AAG-3 ′, 5′-ACA3 ′, and 5′-ATA-3 ′. Reports a sequential selection method leading to domain characterization [Greismanetal., (1997) Science 275 (5300), 657-661, Wolfe et al., (1999) J Mol Biol 285 (5), 1917-1934. ]. The present disclosure uses different methods of selecting zinc finger domains that recognize the site by eliminating target site overlap. First, finger 3 of C7 (RS D- E-RKR) (SEQ ID NO: 3) binding to subsite 5′-GCG-3 ′ was exchanged with a domain containing no aspartic acid at position 2 (FIG. 1). Helix TS G- N-LVR (SEQ ID NO: 6) has previously been characterized to bind to triplet 5′-GAT-3 ′ with high specificity at finger position 2, and appears to be a promising candidate. It was broken. Compare this 3-finger protein (C7.GAT; FIG. 1), which includes the C7 fingers 1 and 2 and the 5′-GAT-3 ′ recognition helix in finger-3, with the original C7 protein (C7.GCG) Multi-target ELISA was used to analyze the DNA binding specificity of targets with different finger-2 subsites. Both proteins bound to the 5′-TGG-3 ′ subsite (however, C7.GCG became 5′-GGG-3 ′ because of the 5 ′ specification of thymine or guanine by Asp ² of finger 3 reported previously. Also combine).

Recognition of the 5 ′ nucleotide of the finger-2 subsite was assessed using a mixture of all 16 5′-XNN-3 ′ target sites (X = adenine, guanine, cytosine or thymine). In fact, the original C7.GCG protein specified guanine or thymine at the 5 ′ position of finger 2, and C7.GAT did not specify a base, so crossover to adenine complementary to 5 ′ thymine. It was shown that the subsite interaction was destroyed. Similar effects have been reported previously for mutants of Zif268 in which Asp ² was replaced by Ala ² by site-directed mutagenesis [Isalan et al., (1997) Proc Natl Acad Sci USA 94 (11). 5617-5621; Dreier et al., (2000) J. Mol. Biol. 303, 489-502]. The affinity of C7.GAT measured by gel mobility displacement analysis has been found to be relatively low, about 400 nM compared to 0.5 nM for C7.GCG [Segal et al., (1999) Proc Natl Acad Sci USA 96 (6), 2758-2863], partly because of the lack of Asp ^{2 in} finger 3.

Based on the 3-finger protein C7.GAT, a library was constructed with the phage display vector pComb3H [Barbas et al., (1991) Proc. Natl. Acad. Sci. USA 88, 7978-7882; Rader et al., ( 1997) Curr. Opin. Biotechnol. 8 (4), 503-508]. Randomization is performed by using the VNS codon doping strategy (V = adenine, cytosine or guanine, N = adenine, cytosine, guanine or thymine, S = cytosine or guanine), -1, 1, 2, It included positions 3, 5 and 6. This could give rise to 24 possibilities for each randomized amino acid position, and the aromatic amino acids Trp, Phe and Tyr and stop codons were excluded from this strategy. Leu since found exclusively from 4 of the recognition helix of _Cys 2 -His ₂ type of zinc finger domains, this position was not randomized. After transformation of the library into ER2537 cells (New England Biolabs), the library contained 1.5 × 10 ⁹ members. This exceeded the required library size by a factor of 60 and was sufficient to contain all combinations of amino acids.

  In the presence of non-biotinylated competitor DNA, six rounds of selection of zinc finger-display phages that each bound to each of the 16 5′-GAT-ANN-GCG-3 ′ biotinylated hairpin target oligonucleotides were performed. The stringency of selection was increased each round by reducing the amount of biotinylated target oligonucleotide and increasing the amount of competitor oligonucleotide mixture. In the sixth round, the target concentration is typically 18 nM, and the 5′-CNN-3 ′, 5′-GNN-3 ′, and 5′-TNN-3 ′ competitor mixtures are each 5-fold excess for each oligonucleotide pool And the specific 5′-ANN-3 ′ mixture (excluding the target sequence) was in a 10-fold excess. Phage that bound to the biotinylated target oligonucleotide were recovered by capture on streptavidin-coated magnetic beads.

Clones were routinely analyzed after the sixth round of selection. The amino acid sequences of the selected finger-2 helices were determined and generally showed good conservation at positions -1 and 3 (Fig. 2), consistent with previously observed amino acid residues at these positions [Segal et al., (1999) Proc Natl Acad Sci USA 96 (6), 2758-2863]. 3 'position -1 when nucleotides are adenine was the Gln, Ser and is selected strongly the 5'-ACA-3' (S PA-D-LTN) ( SEQ ID NO: 77) binding domain was an exception It was. Triplets containing 3 ′ cytosine select Asp ⁻¹ (exceptions were 5′-AGC-3 ′ and 5′-ATC-3 ′ binding domains), 3 ′ guanine is Arg ⁻¹ , and 5 ′ Thymine was Thr- ¹ and His- ¹ . Further, 3 by ^{His -1} 'recognition of thymine, 5'-GAT-3' ( H IS-N-FCR) were observed in the finger 1 of TKK that bind to (SEQ ID NO: 78) [Fairall et al., (1993) Nature (London) 366 (6454), 483-7]. For recognition of intermediate adenine, Asp and Thr were selected at position 3 of the recognition helix. For binding to intermediate cytosine, select Asp ³ or Thr ³ ; for intermediate guanine, His ³ (with the exception of recognition of 5′-AGT-3 ′, which is the unusual amino acid residue His ⁻¹ Therefore, Ser ³ and Ala ³ were selected for intermediate thymine, which may have different binding mechanisms. However, the domain that binds to the 5′-ANG-3 ′ subsite also appears to stabilize the interaction of the 3-finger protein by contacting the 5′-guanine complementary cytosine at the finger-1 subsite. Asp ² is included. Arg and Thr were mainly selected at position 5 of the recognition helix, but positions 1, 2 and 5 were variable.

The most interesting observation was the selection of the amino acid residue at position 6 of the α helix that determines the binding of the 3 bp subsite to the 5 ′ nucleotide. In contrast to recognition of 5 ′ guanine, when direct base contact is achieved by Arg or Lys at position 6 of the helix, direct interaction in the protein / DNA complex for other nucleotides at position 5 ′ is Not observed at all [Elrod-Erickson et al., (1996) Structure 4 (10), 1171-1180; Pavletich et al., (1993) Science (Washington, DC, 1883) 261 (5129), 1701- 7; Kim et al., (1996) Nat Struct Biol 3 (11), 940-945; Fairall et al., (1993) Nature (London) 366 (6454), 483-7; Houbaviy et al., (1996). ) Proc Natl Acad Sci USA 93 (24), 13777-82; Wuttke et al., (1997) J Mol Biol 273 (1), 183-206; Nolte et al., (1998) Proc Natl Acad Sci USA 95 ( 6), 2938-2943]. Selection of the domain for the 5'-GNN-3 'type finger-2 subsite produced a domain containing only Arg ^{6 in} direct contact with 5' guanine [Segal et al., (1999) Proc Natl Acad Sci USA96 (6), 2758-2863]. However, unlike the results for the 5′-GNN-3 ′ zinc finger domain, selection of the phage display library against the 5′-ANN-3 ′ type finger-2 subsite resulted in various amino acid residues: Ala ⁶ , Arg ⁶ , Asn ⁶ , Asp ⁶ , Gln ⁶ , Glu ⁶ , Thr ⁶ or Val ⁶ were identified (FIG. 2). In addition, a domain that recognizes 5′-TAG-3 ′ was selected from this library with the amino acid sequence RED-N-LHT (FIG. 3z) (SEQ ID NO: 71). Thr ⁶ is also present in finger 2 of Zif268 (RSD-H-LT T ) (SEQ ID NO: 79) that binds to 5′-TGG-3 ′, where no direct contact was observed with the Zif268 / DNA complex.

C7. Finger-2 mutant of GAT was subcloned into a bacterial expression vector as a fusion with maltose binding protein (MBP) and the protein was expressed by induction with 1 mM IPTG (protein (p) Give the name of the selected finger-2 subsite). Proteins were tested by solid-phase enzyme immunoassay (ELISA) against each of the 5'-GAT ANN GCG-3 'type 16 finger-2 subsites to determine their DNA binding specificity (Figure 3, black) Bar). In addition, 5′-nucleotide recognition was analyzed by exposing the zinc finger protein to three subsites that differ only in the 5 ′ nucleotide of the specific target oligonucleotide and the intermediate triplet. For example, pAAA was tested on 5′-AAA-3 ′, 5′-CAA-3 ′, 5′-GAA-3 ′, and 5′-TAA-3 ′ subsites (FIG. 3, white bars). . Many of the 3-finger proteins tested showed a sensitive DNA binding specificity for the finger-2 subsite from which they were selected. The binding characteristics of the domains considered to be the most specific binders in each set, boxed in FIG. 2, are represented in the upper panel of FIG. (Marked with an asterisk) is summarized in the lower panel of FIG. The exception was pAGC and pATC whose DNA binding was too weak to be detected by ELISA. PAGC (DAS-H-LHT) containing the predicted amino acids Asp- ¹ and His ³ specifying 3 ′ cytosine and intermediate guanine, and Thr ⁶ not selected in any case for 5 ′ adenine (sequence The most promising helix for number 80) was analyzed without detectable DNA binding.

To analyze a larger set, after a sixth round of selection, a pool of coding sequences for pAGC was subcloned into plasmid pMal and 18 individual clones were tested for DNA binding specificity, all in ELISA. It did not show measurable DNA binding. In the case of pATC, two helices (RRS-S-CRK and RRS-A-CRR) containing the Leu ^{4 to} Cys ⁴ mutation where no DNA binding could be detected (SEQ ID NO: 80, 81) were selected. Since no protein that binds to the finger-2 subsite was generated by phage display, a domain that binds to these 5'-AGC-3 'or 5'-ATC-3' was applied by applying rational design. I found it. 5′-GGC-3 ′ (ERS-K-L AR (SEQ ID NO: 82), DPG-H-L VR (SEQ ID NO: 83)) and 5′-GTC-3 ′ (DPG-A-L VR ) (sequence No. 84) was constructed a finger-2 mutant based on a recognition helix previously demonstrated to bind specifically [Segal et al., (1999) Proc Natl Acad Sci USA 96 (6), 2758- 2763]. For pAGC, two proteins were constructed (ERS-K-L RA (SEQ ID NO: 85)) by simply replacing positions 5 and 6 with 5 ′ adenine recognition motif RA or RV (FIGS. 3a, 3b and 3i). , DPG-H-L RV (SEQ ID NO: 86)). The DNA binding of these proteins was below the detection level. In the case of pATC, two finger-2 mutants (Figure 3b) containing the RV motif (Figure 3b) were constructed (DPG- A- L RV (SEQ ID NO: 87), DPG- S- L RV (sequence Number 88)). Regardless of whether the 3rd position is ALa or Ser, both proteins bound to DNA with very low affinity.

  Analysis of 3-finger proteins for 16 finger-2 subsites by ELISA showed that some finger-2 domains bind best to targets that did not allow them to be selected. First, the helix selected primarily for 5′-AGA-3 ′ is RSD-H-LTN (SEQ ID NO: 63), which in fact bound to 5′-AGG-3 ′ (FIG. 3r). ). This can be explained by Arg in the -1 position. Furthermore, this protein showed a distinct difference in 5 ′ adenine compared to the predominantly selected helix pAGG (RSD-H-LAE (SEQ ID NO: 55); FIG. 3 j). Second, the helix that specifically binds to 5′-AAG-3 ′ (RSD-N-LKN (SEQ ID NO: 61); FIG. 3p) is actually against 5′-AAC-3 ′ (FIG. 2). And in the finger-2 subsite 5′-AAG-3 ′ rather than pAAG (RSD-T-LSN (SEQ ID NO: 48); FIG. 3c), which was selected in the 5′-AAG-3 ′ set. It bound specifically. Furthermore, proteins directed to 5′-ANG-3 ′ type target sites showed cross-reactivity with all 5′-ANG-3 ′ type 4 target sites except for pAGG (FIGS. 3j and 3r). . Also, pAAG (RSD-N-LKN (SEQ ID NO: 61); FIG. 3p) has only weak cross-reactivity, so the recognition of intermediate purines appears to be more limited than in the case of intermediate pyrimidines.

In comparison, the proteins pACG (RTD-T-LRD (SEQ ID NO: 52); FIG. 3 g) and pATG (RRD-A-LNV (SEQ ID NO: 58); FIG. 3 m) were found to be all 5′-ANG-3 ′ subsites. The cross-reactivity of is shown. Recognition of intermediate pyrimidines has been reported to be difficult in previous studies on domains that bind to the 5'-GNG-3 'sequence [Segal et al., (1999) Proc Natl Acad Sci USA 96 (6), 2758-2863; Dreier et al., (2000) J. Mol. Biol. 303, 489-502]. In order to improve the recognition of intermediate nucleotides, a finger-2 mutant containing a different amino acid residue at position 3 was generated by site-directed mutagenesis. pAAG (RSD- T -LSN (SEQ ID NO: 48); Figure 3c) binding ^was more specific for Thr ³ ~Asn 3 mutations after intermediate adenine (Figure 3o). Binding to 5′-ATG-3 ′ (SRD- A- LNV (SEQ ID NO: 58); FIG. 3m) was improved by single amino acid exchange of Ala ^{3 to} Gln ³ (FIG. 3w) and pACG (RSD- T -Thr ^{3 to} Asp ³ or Gln ³ mutations for LRD (SEQ ID NO: 52); FIG. 3g) disrupted DNA binding. Furthermore, the recognition helix pAGT (HRT-T-L L N ( SEQ ID NO: 56); Figure 3k) ^showed cross-reactivity for Leu ⁵ ~Thr 5 is reduced by substitution (Fig. 3s) intermediate nucleotides. Surprisingly, with the improved discrimination for intermediate nucleotides, there was often some loss of specificity for 5 ′ adenine recognition (compare FIGS. 3o-3p, 3m-3w, 3k-3s). .

Binds to 5 ′ adenine or cytosine-containing subsites from the finger-2 library [Segal et al., (1999) Proc Natl Acad Sci USA 96 (6), 2758-2863] based on 3-finger protein C7 Selection of zinc finger domains is not appropriate for selection of zinc finger domains because of the restriction by aspartic acid at position 2 of finger 3 to cross-subsite contact to the 5 ′ complementary nucleotide of the finger-2 subsite (FIG. 1). There wasn't. We have eliminated the contact by exchanging domains and finger 3 lacking Asp ^2. Randomize finger 2 of C7.GAT to construct a phage display library. In most cases, novel 3-finger proteins that bind to the 5'-ANN-3 'type finger-2 subsite were selected. For subsites 5′- A GC-3 ′ and 5′- A TC-3 ′, no rigid binder was identified. Domains that bind to previously selected subsites 5′- G GC-3 ′ and 5′- G TC-3 ′ from C7-based phage display libraries have excellent DNA binding specificity to their target sites This was unexpected since it showed an affinity of 40 nM [Segal et al., (1999) Proc Natl Acad Sci USA 96 (6), 2758-2863]. One simple explanation is a limited randomization strategy through the use of VNS codons that do not contain aromatic amino acid residues. For domains that bind to the 5′-GNN-3 ′ subsite, these were not included in the library because aromatic amino acid residues were not selected even if included in the randomization strategy [Segal et al., (1999) Proc Natl Acad Sci USA 96 (6), 2758-2863]. However, zinc finger domains containing aromatic residues such as finger II of TFIIIA (KN2), eg CFII2 (VKD- Y- LTK (SEQ ID NO: 89); [Gogos et al., (1996) PNAS 93, 2159-2164]). W -K-LQA (SEQ ID NO: 90); [. Wuttke et al , (1997) J Mol Biol 273 (1), 183-206] finger 1, TTK in) (HIS-N- F CR (SEQ ID NO: 78) [Fairall et al., (1993) Nature (London) 366 (6454), 483-7])) finger 1 and GLI (AQ Y -M-LVV (SEQ ID NO: 91); [Pavletich et al., (1993) The finger 2 of Science (Washington, DC, 1883-) 261 (5129), 1701-7]) has been reported so far. Aromatic amino acid residues may be important for recognition of subsites 5′-AGC-3 ′ and 5′-ATC-3 ′.

Recently, it has become apparent that the recognition helix of the Cys ₂ -His ₂ zinc finger domain can take different orientations on DNA to achieve optimal binding [Pabo et al., (2000) J. Mol. Biol. 301, 597-624]. However, the orientation of the helix in this region can be limited in part by frequently observed interactions involving zinc ions, His ⁷ and the phosphate backbone. Furthermore, when comparing the binding properties of the interactions in the protein / DNA complex, the C-α atom at position 6 is usually 8.8 ± 0.8 angstroms away from the nearest heavy atom of the 5 ′ nucleotide in the DNA subsite. It was concluded that this is only advantageous for the recognition of 5 ′ guanine by Arg ⁶ or Lys ⁶ [Pabo et al., (2000) J. Mol. Biol. 301, 597-624]. To date, no interaction between other 6-position residues and bases other than guanine has been observed from protein / DNA complexes. For example, the fingers 4 of YY1 (QST-N-LK S ) ( SEQ ID NO: 92), 5'C AA-3 'recognizes that, Ser ⁶ and 5' contact between the cytosine was observed [ Houbaviy et al., (1996) Proc Natl Acad Sci USA 93 (24), 13577-82]. Further, Zif268 (specifying finger 3 of YY1 (LDF-N-LR T ) (SEQ ID NO: 93), which recognizes 5′- A TT-3 ′, and 5′- T / G GG-3 ′ In the case of Thr ⁶ in finger 2 of (RSD-H-LT T ) (SEQ ID NO: 79), no contact with the 5 ′ nucleotide was observed [Houbaviy et al., (1996) Proc Natl Acad Sci USA 93 (24 ), 13577-82; Elrod-Erickson et al., (1996) Structure 4 (10), 1171-1180]. Finally, Ala ⁶ of finger 2 of the tram track (RKD-N-MT A ) (SEQ ID NO: 94) that binds to subsite 5′- A AG-3 ′ does not contact 5 ′ adenine [Fairall et al (1993) Nature (London) 366 (6454), 483-7].

It has been demonstrated to bind effectively to 5 ′ guanine in different situations [Segal et al., (1999) Proc Natl Acad Sci USA 96 (6), 2758-2663], amino acid residues Ala ⁶ , Val ⁶ , Asn. ⁶ and up to Arg ⁶ were mainly selected from the C7.GAT library for the 5′-ANN-3 ′ type (FIG. 2) DNA subsite. In addition, position 6 was selected as Thr, Glu and Asp by the finger-2 target site. This is consistent with earlier studies by other groups in which the proximity finger positions were randomized [Jamieson et al., (1996) Proc Natl Acad Sci USA 93, 12834-12839; Isalan et al., (1998). Biochemistry 37 (35), 12026-12033]. As a result of screening of the phage display library, the same Tyr, Val, Thr, Asn, Lys, Glu and Leu, and Gly, Ser and Arg were selected for the recognition of 5 ′ adenine instead of the amino acid residue Ala. In addition, by using a sequential phage display selection strategy, several domains that bind to the 5′-ANN-3 ′ subsite were identified, and specificity was assessed by target site selection. Arg, Ala, and Thr at position 6 of the helix have been demonstrated to recognize primarily 5 'adenine [Wolfe et al., (1999) Annu. Rev. Biophys. Biomol. Struct. 3, 183-212]. .

Furthermore, as indicated by target site selection for finger 5 of Gfi-1 (QSS-N-LI T ) (SEQ ID NO: 95) binding to subsite 5′-AAA-3 ′ [Zweidler-McKay et al ., (1996) Mol. Cell. Biol. 16 (8), 4024-4034] Thr ⁶ specifies 5 'adenine. These examples, including this result, indicate that a relationship exists between the 6th amino acid residue and the 5 ′ adenine because they are frequently selected. This is inconsistent with data obtained from crystallographic studies, which never showed an interaction at the 6-position of the _-helix with 5 'nucleotides other than guanine. As one simple explanation, short amino acid residues, such as Ala, Val, Thr or Asn, may not be steric hindrance in the binding mode of the domain that recognizes the 5′-ANN-3 ′ subsite. This is evidenced by the results gathered by site-directed mutagenesis at position 6 for the helix (QRS-A-LT V ) (SEQ ID NO: 96) that binds to the 5′-G / ATA-3 ′ subsite. [Gogos et al., (1996) PNAS 93, 2159-2164]. Substituting Val ⁶ for Ala ⁶ found also for the domains described herein had no effect on binding specificity or affinity.

By using computer modeling, the possible interaction of 5 ′ adenine with frequently selected Ala ⁶ , Asn ⁶ and Arg ⁶ was investigated. Analysis of the interaction from Ala ⁶ in the helix that binds to 5 ′ adenine and 5′-AAA-3 ′ (QRA-N-LR A ; FIG. 3a) (SEQ ID NO: 46) showed that finger 1 from the Zif268 mutant ( QSG-S-LTR) (SEQ ID NO: 97) based on the coordinates of the protein / DNA complex. QRA-N-LRA (SEQ ID NO: 98) ^{If Gln -1} and Asn ³ bind their hydrogen and the adenine base in a defined way, by these interactions, the Ala ⁶ between methyl and 5 'adenine It also suggests that direct contact cannot be proposed for Val ⁶ and Thr ⁶ since the distance beyond 11 angstroms between the thymine base paired with the methyl group of Ala ⁶ and Ala ⁶ and adenine is naturally fixed. Has been.

Interestingly, the predicted lack of 5 ′ specificity due to the short amino acid at position 6 of the α-helix is only partially verified by the binding data. Helixes such as RRD-A-LN V (SEQ ID NO: 58) (FIG. 3m) and C7.GAT finger-2 helix RSD-H-LT T (SEQ ID NO: 5) are in fact essentially 5 ′ specific. Did not show. However, helix DSG-N-LR V (SEQ ID NO: 47) (FIG. 3b) shows excellent specificity for 5 ′ adenine and TSH-G-LT T (SEQ ID NO: 70) (FIG. 3y) is 5 ′ adenine. Or specific for guanine. Other helices with a short 6-residue showed varying degrees of 5 'specificity, and the only apparent consistency was that 5' thymine was normally excluded (Figure 3). The observed binding pattern should be derived from another source, since it is unlikely that the residue at position 6 can contribute directly to specificity. Possibilities include localized sequence-specific DNA structures and overlapping interactions from neighboring domains. However, the residue at position 2 of finger 3 (often observed to contact a neighboring site) is a glycine in the parent protein C7.GAT, and 5 'thymine is eliminated by the two helices described above. The latter possibility is inconvenient because it was not done.

Asparagine was also often selected in sixth place. Helix HRT-T-LT N (SEQ ID NO: 56) (FIG. 3k) and RSD-T-LS N (SEQ ID NO: 48) (FIG. 3c) showed excellent specificity for 5 ′ adenine. However, Asn ⁶ also appears to confer specificity for both adenine and guanine (Figures 3n, 3p and 3r), suggesting an interaction with N7 common to both nucleotides. In Zif268 / DNA crystal structure (RSD-H-LT T (SEQ ID NO: 79); [Elrod-Erickson et al., (1996) Structure 4 (10), 1171-1180]), 5′-TGG-3 ′ Computer modeling of the helix binding to 5′-AGG-3 ′ (RSD-H-LT N (SEQ ID NO: 90); FIG. 3r), based on the coordinates of finger 2 to bind, shows that N-δ of Asn ⁶ is It was suggested to be about 4.5 angstroms from N7 of 5 ′ adenine. Possible moderate reorientation of the helix within the prescribed docking orientation [Pabo et al., (2000) J. Mol. Biol. 301, 597-624], glutamate instead of arginine is in the -1 position Similar to the reorientation observed when it appears, N-δ can be introduced in a reasonable manner within the hydrogen bond spacing. However, it is interesting to speculate why Asn ⁶ was selected in this 5′-ANN-3 ′ recognition set and the longer Gln ⁶ was not selected. Gln ⁶ is more variable and may have been able to stabilize other interactions selected during phage display. Alternatively, the shorter side chain of Asn ⁶ can accommodate an ordered water molecule that can contact 5 ′ nucleotides without helix reorientation.

The last residue to consider is Arg ⁶ . In our previous study, it was selected without objection to recognize 5 'guanine with high specificity [Segal et al., (1999) Proc Natl Acad Sci USA 96 (6), 2758-2863], Arg ⁶ It was somewhat surprising that is selected very frequently with the 5′-ANN-3 ′ target. However, in recent studies, in some cases, in addition to recognition of 5 ′ guanine (FIGS. 3t and u), Arg ⁶ mainly identified 5 ′ adenine (FIGS. 3e, f, h and v). Based on the coordinates of Zif268 mutant finger 1 QSG-S-LTR (SEQ ID NO: 98) binding to 5′-GCA-3 ′ [Elrod-Erickson et al., (1998) Structure 6 (4), 451- 464] Computer modeling of the helix binding to 5′-ACA-3 ′ (SPA-D-LT R (SEQ ID NO: 50); FIG. 3e) shows that Arg ⁶ is O4 of thymine base-paired to 5 ′ adenine. This suggests that the conformation that enables cross-chain hydrogen bonding with can be easily adopted. In fact, Arg ⁶ can bind with both thymine O4 and guanine O6 base paired with intermediate cytosine with a good binding structure. Such an interaction is consistent with the fact that Arg ⁶ was selected almost without objection when the target sequence was 5'-ACN-3 '. The expectation that arginine promotes multiple interactions is inevitable. Several lysines in TFIIIA have been observed by NMR to be conformationally variable [Foster et al., (1997) Nat. Struct. Biol. 4 (8), 605-608], Gln- ¹ behaves in a manner that suggests variability [Dreier et al., (2000) J. Mol. Biol. 303, 489-502]. Arginine has a rotatable bond and higher hydrogen bond potential than lysine or glutamine, and the prediction that Arg ⁶ is not limited to recognition of 5 ′ guanine is interesting.

The amino acid residues at positions -1 and 3 were generally selected in a manner similar to their 5'-GNN-3 'counterparts with two exceptions. His- ¹ was selected for pAGT and pATT and recognized 3 ′ thymine (FIGS. 3k, 3n and 3y) and Ser- ¹ was selected for pACA and recognized 3 ′ adenine (FIGS. 3e and 3t). Gln ^-1 is often used to specify 3 'adenine in 5'GNN-3' type subsites, and a new component of 3 'adenine recognition forms hydrogen bonds with 3' adenine This study suggested that it contains Ser- ¹ selected for a domain that recognizes a possible 5′-ACA-3 ′ subsite (FIG. 2). Computer modeling shows van der Waals contact with the methyl group of thymine where Ala ² selected together in the helix SP A -D-LTR (SEQ ID NO: 50) (Figure 3e) is potentially base paired with 3 'adenine. It is proved that you can have. The best evidence that Ala ² can be included is that helix SP A -D-LTR (SEQ ID NO: 50) (Figure 3e) shows strong specificity for 3 'adenine and SH S -D-LVR (SEQ ID NO: 65) (FIG. 3t) is not shown. Gln- ¹ is often sufficient for 3 ′ adenine recognition. However, data from our previous studies suggested that the side chain of Gln- ¹ may adopt multiple conformations that allow recognition of, for example, 3 ′ thymine [Nardelli et al., (1992) Nucleic Acids Res. 20 (16), 4137-44; Elrod-Erickson et al., (1998) Structure 6 (4), 451-464; Dreier et al., (2000) J. Mol. Biol. 303, 489-502]. Ala ^{2 in} combination with Ser- ¹ may be an alternative means of specifying 3 ′ adenine.

Another interaction not observed in the 5′-GNN-3 ′ test is the cooperative recognition of 3 ′ thymine by residues at His- ¹ and position 2. In the finger 1 of the crystal structure of tram tracks / DNA complex, the helix H IS-N-FCR (SEQ ID NO: 99) binds to the subsite 5'-GAT-3 '[Fairall et al., (1993) Nature (London) 366 (6454), 483-7]. The His ^-1 ring is perpendicular to the plane of the 3 'thymine base and is about 4 angstroms from the methyl group. Ser ² also forms hydrogen bonds with O 4 of 3 ′ thymine. A similar series of contacts can be foreseen by computer modeling for the recognition of 5'-ATT-3 'by helix HKN-A-LQN (SEQ ID NO: 100) (Figure 3n). Asn ² in this helix has the potential to hydrogen bond not only with 3 ′ thymine but also with adenine base paired with thymine. His ^-1 also in combination with ^{Thr 2 5'-AGT-3 '} (H RT-T-LLN ( SEQ ID NO: 98); Figure 3k) it was also found for the helix that binds. Thr is structurally similar to Ser and may be involved in similar recognition mechanisms.

In conclusion, the characterization of the zinc finger domain reported in this study that binds to the 5'-ANN-3 'DNA subsite is consistent with the general view that there is no universal recognition code, and therefore The rational design of additional domains has become difficult. However, phage display selection can be applied and pre-specified zinc finger domains can serve as modules for constructing artificial transcription factors. Here characterization domain is 5 - to allow targeting of '(GNN) ₆ -3' other DNA sequences. G / C rich sequences are often comprising a binding site for cell proteins, 5 for 'the (GNN) ₆ -3' sequence is not necessarily obtained found from every promoter, which is an important supplement to the existing domain It is.

II. Polynucleotides, expression vectors and transformed cells The present invention encompasses nucleotide sequences that encode zinc finger-nucleotide binding polypeptides. DNA sequences encoding the zinc finger-nucleotide binding polypeptides of the invention can be obtained by several methods, including native, truncated and expanded polypeptides. For example, DNA can be isolated using hybridization procedures known in the art. These include (1) hybridization of genomic or cDNA libraries and probes to detect shared nucleotide sequences, (2) antibody screening of expression libraries to detect covalent structural features, and (3) There is, but is not limited to, synthesis by polymerase chain reaction (PCR). The RNA sequences of the present invention can be obtained by methods known in the art (see, eg, Current Protocols in Molecular Biology, Ausubel, et al. Ed. (1989)).

  The development of specific DNAs encoding the zinc finger-nucleotide binding polypeptides of the present invention is based on (1) isolation of double stranded DNA sequences from genomic DNA, (2) chemical production of DNA sequences, It can be achieved by providing codons necessary for the polypeptide of interest and (3) in vitro synthesis of double stranded DNA sequences by reverse transcription of mRNA isolated from eukaryotic donor cells. In the latter case, a double-stranded DNA complement of mRNA is eventually formed, which is commonly referred to as cDNA. Of these three methods of developing specific DNA sequences for use in recombinant techniques, isolation of genomic DNA is the rarest. This is especially true when it is desired to achieve microbial expression of the mammalian polypeptide due to the presence of introns.

  To obtain a zinc finger-derived DNA binding polypeptide, synthesis of a DNA sequence is often a selective method when the entire sequence of amino acid residues of the desired polypeptide product is known. When the entire sequence of amino acid residues of the desired polypeptide is not known, direct synthesis of the DNA sequence is not possible and the method of selection is the formation of a cDNA sequence. A standard procedure for isolating a cDNA sequence of interest is the formation of a plasmid-carrying cDNA library derived from reverse transcription of mRNA abundant in donor cells that exhibit high levels of gene expression. When used in combination with the polymerase chain reaction technique, even rare expression products can be cloned. If a significant portion of the amino acid sequence of a polypeptide is known, the production of labeled single- or double-stranded DNA or RNA probe sequences that replicate the presumed sequence present in the target cDNA is denatured into single-stranded form And can be used in DNA / DNA hybridization procedures performed on cloned copies of cDNA (Jay et al., Nucleic Acid Research 11: 2325, 1983).

  The polypeptides of this invention can be operably linked to one or more functional peptides. Such functional peptides are known in the art and can be transcriptional regulators, such as repressors or activation domains or peptides with other functions. Exemplary preferred functional peptides are nucleases, methylases, nuclear localization domains, and restriction enzymes such as endo- or ectonucleases (eg, Chandrasegaran and Smith, Biol. Chem., 380: 841-848). (1999)).

  An exemplary repression domain peptide is an ERF repressor domain (ERD) identified by amino acids 473-530 of ets2 repressor factor (ERF) (Sgouras, DN, Athanasiou, MA, Beal, GJ, Jr., Fisher, RJ, Blair, DG and Mavrothalassitis, GJ (1995) EMBO J. 14, 4781-4793). This domain transmits the antagonism of ERF to the activity of transcription factors of the ets family. A synthetic repressor is constructed by fusion of this domain to the N- or C-terminus of the zinc finger protein. The second repressor protein is produced using the Kruppel associated box (KRAB) domain (Margolin, JF, Friedman, JR, Meyer, W., K.-H., Vissing, H., Thiesen, H.- J. and Rauscher III, FJ (1994) Proc. Natl. Acad. Sci. USA 91, 4509-4513). This repressor domain is generally found at the N-terminus of the zinc finger protein and is RING finger protein KAP-1 (Friedman, JR, Fredericks, WJ, Jensen, DE, Speicher, DW, Huang, X.-P., Neilson, EG & Rauscher III, FJ (1996) Genes & Dev. 10, 2067-2078) appears to exert its repressive activity on TATA-dependent transcription in a distance- and orientation-independent manner (Pengue, G. & Lania, L. (1996) Proc. Natl. Acad. Sci. USA 93, 1015-1020). We have developed zinc finger protein KOX1 (Margolin, JF, Friedman, JR, Meyer, W., K.-H., Vissing, H., Thiesen, H.-J. and Rauscher III, FJ (1994) Proc. Natl. Acad. Sci. USA 91, 4509-4513), the KRAB domain found between amino acids 1 and 97 was used. In this case, an N-terminal fusion with the zinc finger polypeptide is constructed. Finally, to investigate the usefulness of histone deacetylation for inhibition, amino acids 1-36 of Mad mSIN3 interaction domain (SID) are fused to the N-terminus of zinc finger protein (Ayer, DE, Laherty, CD, Lawrence, QA, Armstrong, AP & Eisenman, RN (1996) Mol. Cell. Biol. 16, 5772-5781). This small domain is found at the N-terminus of the transcription factor Mad and is involved in the transmission of its transcriptional repression by interaction with mSIN3, which in turn interacts with the co-repressor N-CoR and histone deacetylase mRPD1. (Heinzel, T., Lavinsky, RM, Mullen, T.-M., Ssderstrsm, M., Laherty, CD, Torchia, J., Yang, W.-M., Brard, G., Ngo, SD & al., e. (1997) Nature 387, 43-46). To investigate gene-specific activation, amino acids 413-489 of herpes simplex virus VP16 protein (Sadowski, I., Ma, J., Triezenberg, S. & Ptashne, M. (1988) Nature 335, 563-564), Or an artificial tetramer repeat of the minimal activation domain of VP16 (Seipel, K., Georgiev, O. & Schaffner, W. (1992) EMBO J. 11, 4961-4968), termed VP64. By fusing the finger polypeptide, a transcriptional activator is generated.

III. Pharmaceutical Compositions In another aspect, the invention combines a therapeutically effective amount of a zinc finger-nucleotide binding polypeptide or a therapeutically effective amount of a nucleotide sequence encoding a zinc finger-nucleotide binding polypeptide with a pharmaceutically acceptable carrier. Pharmaceutical compositions comprising the combination are provided.

  As used herein, the terms “pharmaceutically acceptable”, “physiologically acceptable” and their grammatical variations are interchangeable when referring to compositions, carriers, diluents and reagents. Used to indicate that these substances can be administered to humans without causing undesired physiological effects such as nausea, dizziness, stomach upset and the like that prohibit the administration of the composition.

  The preparation of a pharmacological composition that contains active ingredients dissolved or dispersed therein is well understood in the art. Typically, the composition is prepared as a liquid solution or suspension as a sterile injectable material, however, suitable aqueous or non-aqueous to be in solution or suspension in liquid prior to use. Solid forms can also be produced. The preparation can also be emulsified.

  The active ingredient can be mixed with excipients that are pharmaceutically acceptable and compatible with the active ingredient and in amounts suitable for use in the therapeutic methods described herein. Suitable excipients are, for example, water, saline, dextrose, glycerin, ethanol and the like and combinations thereof. In addition, if desired, the composition may contain minor amounts of auxiliary substances such as wetting or emulsifying agents, pH buffering agents and the like which enhance the effectiveness of the active ingredient.

  The therapeutic pharmaceutical composition of the present invention may include pharmaceutically acceptable salts of the components therein. Pharmaceutically acceptable salts include acid addition salts (formed by the free amino group of the polypeptide) formed with inorganic acids such as hydrochloric acid or phosphoric acid, or organic acids such as acetic acid, tartaric acid, mandelic acid, and the like. To do. Salts formed by free carboxyl groups are also derived from inorganic bases such as sodium, potassium, ammonium, calcium or iron hydroxide, and organic bases such as isopropylamine, trimethylamine, 2-ethylaminoethanol, histidine, procaine, etc. Can be done.

  Physiologically acceptable carriers are known in the art. Examples of liquid carriers are sterile aqueous solutions that do not contain active ingredients and substances other than water, or that contain a buffer, such as sodium phosphate at physiological pH values, saline, or both, such as phosphate buffered saline. There is. Still further, the aqueous carrier may contain a plurality of buffer salts and salts such as sodium chloride and potassium, dextrose, propylene glycol, polyethylene glycol and other solutes. The liquid composition may also include a liquid phase in addition to and excluding water. Examples of such additional liquid phases are glycerin, vegetable oils such as cottonseed oil, organic esters such as ethyl oleate, and water-oil emulsions.

IV. Use In one aspect, the method of the invention modulates (inhibits or suppresses) the expression of a nucleotide sequence comprising a zinc finger-nucleotide binding motif, wherein the zinc finger-nucleotide binding motif binds to the motif. A method comprising contacting with an effective amount of a finger-nucleotide binding polypeptide. If the nucleotide sequence is a promoter, this method inhibits transcriptional activation (transactivation) of the promoter containing the zinc finger-DNA binding motif. The term “inhibit” refers to suppression of the level of transcriptional activation of a structural gene operably linked to a promoter, including, for example, a zinc finger-nucleotide binding motif. Furthermore, zinc finger-nucleotide binding polypeptide derivatives can bind to motifs within structural genes or RNA sequences.

  The term “effective amount” refers to an amount that results in inactivation of a previously activated promoter or an amount that results in inactivation of a promoter containing a zinc finger-nucleotide binding motif, or blocks transcription of a structural gene or translation of RNA The amount to be included. The amount of zinc finger-derived nucleotide binding polypeptide required is the amount required to replace the natural zinc finger-nucleotide binding protein in the existing protein / promoter complex or competes with the natural zinc finger-nucleotide binding protein. Thus, it is an amount necessary to form a complex with the promoter itself. Similarly, the amount required to block a structural gene or RNA is the amount that binds to RNA polymerase and blocks it from reading through the gene or inhibits translation, respectively. Preferably, the method is performed intracellularly. Transcription or translation is suppressed by functionally inactivating the promoter or structural gene. Delivery of an effective amount of an inhibitory protein to bind or “contact” a cellular nucleotide sequence comprising a zinc finger-nucleotide binding protein motif is one of the mechanisms described herein, such as a retroviral vector or liposome, or It can be achieved by other methods known in the art.

  The term “modulate” refers to the suppression, promotion or induction of function. For example, the zinc finger-nucleotide binding polypeptides of the present invention modulate a promoter sequence by binding to a motif within the promoter, thereby facilitating or repressing transcription of a gene operably linked to the promoter nucleotide sequence. Can do. Alternatively, modulation can involve inhibition of gene transcription, in which case zinc finger-nucleotide binding polypeptides bind to a structural gene and block DNA-dependent RNA polymerase from reading through the gene. , Gene transcription is inhibited. The structural gene can be a normal cellular gene or for example an oncogene. Alternatively, modulation can include inhibition of transcript translation.

  The promoter region of a gene typically includes regulatory elements that are 5 'to the structural gene. When the gene is to be activated, a protein known as a transcription factor binds to the promoter region of the gene. This combination is similar to an “on switch” by allowing the enzyme to transcribe the second gene segment from DNA to RNA. In most cases, the generated RNA molecule serves as a template for specific protein synthesis, and the RNA itself may be the final product.

  The promoter region can be a normal cellular promoter or for example an onco-promoter. Onco-promoter is generally a virus-derived promoter. For example, the retroviral long terminal repeat (LTR) is a promoter region that may be a target for the zinc finger binding polypeptide variants of the present invention. Promoters from members of the retrovirus group, including pathogens such as human T cell tropic virus (HTLV) 1 and 2, or human immunodeficiency virus (HIV) 1 or 2, are transcribed by the zinc finger binding polypeptides of the invention. FIG. 4 is an example of a viral promoter region that can be targeted for modulation.

To investigate whether a domain described herein that specifically binds to a 5′-ANN-3 ′ DNA sequence is appropriate for the construction of the artificial transcription factor, various numbers of 5′-ANN-3 'Four 6-finger proteins containing domains were assembled. For each of the 6-finger proteins, two 3-finger coding regions were generated by PCR overlap extension using the Sp1C framework [Beerli et al., (1998) Proc Natl Acad Sci USA 95 (25), 14628-14633. ]. These 6-finger proteins were then fused to create 6-finger proteins via restriction sites (FIG. 4a) and cloned into the bacterial expression vector pMal for analysis of DNA binding specificity and affinity. . First, designed to recognize the indefinite 18 bp target site 5′-ATG-TAG-AGA-AAA-ACC-AGG-3 ′, completely free of 5′-GNN-3 ′ triplets, The finger protein pAart was constructed. Second, three 6-finger proteins were constructed that contained both 5'-GNN-3 'and 5'-ANN-3' domains. Previously we, the regulation of endogenous genes, 5 - showed '(GNN) ₆ -3' are attached to the DNA sequence, the 6-finger protein pE2C or pE3 be specifically achieved each erbB- A well-characterized model of the 2 and erbB-3 genes [Beerli et al., (2000) Proc Natl Acad Sci USA 97 (4), 1495-1500; Beerli et al., (2000) J. Biol. .275 (42), 32617-32627] was selected for testing.

6-finger that binds to the target site 5′- A CC GGA GAA A CC A GG GGA-3 ′ (SEQ ID NO: 101) located at positions −168 to −151 of the 5 ′ untranslated region (UTR) of the erbB-2 gene The protein pE2X was constructed (Figure 4a). In addition, two proteins linked by the 5′UTR of the erbB-3 gene were generated. The protein pE3Y binds to the target site 5′- A TC GAG GCA A GA GCC A CC-3 ′ (SEQ ID NO: 102) at positions −94 to −111 of the 5′UTR and is at positions −79 to −61 pE3Z recognizes 5′-GCC GCA GCA GCC A CC A AT-3 ′ (SEQ ID NO: 103) (FIG. 4a). The coding sequences for the four 6-finger proteins were then cloned into the bacterial expression vector pMal. Crude extracts containing zinc finger-MBP fusion protein were tested for DNA binding in ELISA (FIG. 4b). All four proteins showed a sharp binding specificity to their target DNA, with no cross-reactivity with other target sites tested. Affinity was measured by gel mobility displacement assay with purified protein. The protein Aart binds to its DNA target site with an affinity of 7.5 pM, pE2X binds with an affinity of 15 nM, pE3Y binds with 8 nM and pE3Z with 2 nM, which we have analyzed so far. It is included in the range of affinity observed for most 6-finger proteins.

  To assess the potential for specific gene regulation, the protein coding sequence for Aart was cloned into the vector pcDNA and the VP64 activation domain, the herpes simplex virus protein VP16 [Seipel et al., (1992) EMBO J. 11 (13), 4961-4968; Beerli et al., (1998) Proc Natl Acad Sci USA 95 (25), 14628-14633] fused to a tetrameric repeat of the minimal activation domain. HeLa cells were transiently co-transfected with the effector construct coding for the zinc finger protein only or as a fusion with the VP64 domain, and a luciferase reporter plasmid under the control of a minimal promoter containing a zinc finger binding site and a TATA box. The Aart binding site was present in 5 copies and the promoter used as a control contained 6 2C7 binding sites. Luciferase expression was up-regulated 2000-fold by the pAart-VP64 fusion protein compared to the control without the activation domain (FIG. 5a). Activation was specific since no modulation of the reporter containing the 6x2C7 binding site was observed (Figure 5b). As an additional control for specificity, the 6-finger protein p2C7 [Wu et al., (1995) PNAS 92, 344-348] was also tested and only when the promoter contains a 6 × 2C7 binding site (FIG. 5b). Luciferase expression was activated but not when the promoter contained 5 × Aart binding (FIG. 5a). The 3-finger protein in each half of pAart fused to VP64 is unable to activate luciferase expression, which is consistent with previous results [Beerli et al., (2000) Proc Natl Acad Sci USA 97 (4), 1495-1500; Beerli et al., (2000) J. Biol. Chem. 275 (42), 32617-32627].

  To test that the 6-finger proteins pE2X, pE3Y and pE3Z can transcriptionally regulate the endogenous erbB-2 and erbB-3 genes respectively, the coding sequence was subcloned into the retroviral vector pMX-IRES-GFP and Fusion to VP64 activation or KRAB repression domain [Margolin et al., (1994) Proc. Natl. Acad. Sci. USA 91, 4509-4513; Beerli et al., (1998) Proc Natl Acad Sci USA 95 (25 ) 14628-14633]. A retrovirus was used to infect the human cancer cell line A431. Three days after infection, cells were subjected to flow cytometry to analyze the expression levels of ErbB-2 and ErbB-3 (FIG. 6). Infection efficiency was measured by measuring GFP expression. All cell pools were infected at over 80%, except for pE2X-VP64. To measure the expression levels of ErbB-2 and ErbB-3, cells were stained with a specific antibody or a control antibody specific for ErbB-1. The fusion protein pE2X-VP64 was able to upregulate ErbB-2 expression, but only 50% of the cells, which may be due to low infection efficiency. pE3Y was as effective as previously reported pE3 because it showed specific up- and down-regulation when fused to VP64 or KRAB, respectively. Although pE3Z has the highest affinity among the three generated proteins, the pE3Z fusion protein did not alter erbB-3 gene expression. The zinc finger domains described herein that specifically recognize 5'-ANN-3 'DNA sequences are based on the number of DNA sequences that can be targeted by 6-finger proteins and zinc finger-based transcription factors that can currently be constructed. A big contribution.

Example 1: Construction of zinc finger library and selection by phage display The construction of the zinc finger library was based on the previously described C7 protein ([Wu et al., (1995) PNAS 92, 344- 348]; FIG. 1). Finger 3 that recognizes the 5′-GCG-3 ′ subsite is identified as a primer coding for finger 3 (5′-GAGGAAGTTTGCCCACCAGTGGCACACCTGGTGAGGCATACCAAAATC-3 ′) (SEQ ID NO: 104) and pMal-specific primer (5′-GTAAAACGACGCGCGCGTGCCAAGC-3) With a domain that binds to the 5'-GAT-3 'subsite [Segal et al., (1999) Proc Natl Acad Sci USA 96 (6), 2758-2863] using an overlapping PCR strategy using (SEQ ID NO: 105). Replaced. Randomization of zinc finger libraries by PCR overlap extension was essentially as reported [Wu et al., (1995) PNAS 92, 344-348; Segal et al., (1999) Proc Natl Acad Sci USA 96 (6), 2758-2863]. The library was ligated to the phagemid vector pComb3H [Rader et al., (1997) Curr. Opin. Biotechnol. 8 (4), 503-508]. Previously reported procedures [Barbas et al., (1991) Methods: Companion Methods Enzymol. 2 (2), 119-124; Barbas et al., (1991) Proc. Natl. Acad. Sci. USA 88, 7978. -7982; Segal et al., (1999) Proc Natl Acad Sci USA 96 (6), 2758-2863], phage growth and precipitation. Volume 500μl of zinc Buffer A (ZBA: 10 mM Tris, pH 7.5 / ZnCl ₂ in KCl / 1 mM of MgCl _2/90 [mu] M of 90 mM) /0.2PasentoBSA/5mM of DTT / 1% blot (Bio-Rad) The binding reaction was performed in 20 μg of double-stranded sheared herring semen DNA containing / 100 μl precipitated phage (10 ¹³ colony forming units). The phage was allowed to bind to the non-biotinylated competitor oligonucleotide for 1 hour at 4 ° C. before adding the biotinylated target oligonucleotide. Binding continued overnight at 4 ° C. After incubation with 50 μl streptavidin-coated magnetic beads for 1 hour (dynal; blocked by 5% blot in ZBA), the beads were buffered 10 times with 500 μl ZBA / 2% Tween 20/5 mM DTT and Tween free buffer. Washed once with liquid. The elution of bound phage was performed by incubation with 25 μl trypsin (10 μg / ml) in TBS (Tris-buffered saline) for 30 minutes at room temperature. The hairpin competitor oligonucleotide has the sequence 5′-GGCCGCN′N′N′ATC GAGTTTCTCGATNNNGCGGCC-3 ′ (SEQ ID NO: 106), where NNN represents the finger-2 subsite oligonucleotide, N′N′N ′ (Representing the complementary base) (the target oligonucleotide was biotinylated). Target oligonucleotides were usually added at 72 nM in the first three rounds of selection and then reduced to 36 nM and 18 nM in the sixth and final rounds. The 5'-TGG-3 'finger-2 subsite oligonucleotide was used as a competitor to compete with the parent clone. An equimolar mixture of 15 finger-2,5′-ANN-3 ′ subsites, excluding the target site, and 5′-CNN-3 ′, 5′-GNN-3 ′, and 5′-TNN-3 ′ A competitor mixture of each finger-2 subsite of the type was added in increasing amounts with each successive selection round. Typically, no specific 5′-ANN-3 ′ competitor mixture was added in the first round.

  Multiple Target Specificity Assay and Gel Mobility Displacement Analysis—Zinc finger coding sequences were subcloned from pComb3H into a modified bacterial expression vector pMal-c2 (New England Biolabs). After transformation into XL1-Blue (Stratagene), 1 nM isopropyl β-D-thiogalactoside (IPTG) was added followed by expression of zinc finger-maltose-binding protein (MBP) fusion. Frozen / thawed extracts of these bacterial cultures were applied at a 1: 2 dilution to streptavidin-coated 96-well plates (Pierce), each of 16 5′-GAT ANN-GCG-3 ′ target sites. Each was tested for DNA binding specificity. Essentially as reported [Segal et al., (1999) Proc Natl Acad Sci USA 96 (6), 2758-2863; Dreier et al., (2000) J. Mol. Biol. 303, 489-502]. An ELISA (solid phase enzyme linked immunoassay) was performed. After incubation with mouse anti-MBP (maltose binding protein) antibody (Sigma, 1: 1000), a goat anti-mouse antibody coupled with alkaline phosphatase (Sigma, 1: 1000) was applied. After detection, alkaline phosphatase substrate (Sigma) was added and OD405 was measured with SOFTMAX 2.35 (Molecular Devices).

  Gel displacement analysis was performed on purified protein (protein fusion and purification system, New England Biolabs) essentially as reported.

Example 2: Site-directed mutagenesis of finger 2 The reported procedure [Segal et al., (1999) Proc Natl Acad Sci USA 96 (6), 2758-2863; Dreier et al., (2000) J. Mol. Biol. 303, 489-502], a finger-2 mutant was constructed by PCR. As a PCR template, a library clone containing 5′-TGG-3 ′ finger 2 and 5′-GAT-3 ′ finger 3 was used. The PCR product containing mutant finger 2 and 5'-GAT-3 'finger 3 was subcloned into the modified pMal-c2 vector (New England Biolabs) via the NsiI and SpeI restriction sites in frame with C7 finger 1. .

  Polydactylzinc finger protein construction-In a reported manner [Beerli et al., (1998) Proc Natl Acad Sci USA 95 (25), 14628-14633], a finger-2 stitching method using the SP1C framework was used to generate 3 -A finger protein was constructed. The protein produced in this step is a helix that recognizes the 5′-GNN-3 ′ DNA sequence [Segal et al., (1999) Proc Natl Acad Sci USA 96 (6), 2758-2863], as well as described herein. Of 5′-ANN-3 ′ and 5′-TAG-3 ′ helices. Six finger proteins were assembled via compatible XmaI and BsrFI restriction sites. Analysis of DNA binding properties was performed from IPTG-induced frozen / thawed bacterial extracts. To analyze the ability of these proteins to regulate gene expression, the previously reported procedure ([Beerli et al., (1998) Proc Natl Acad Sci USA 95 (25), 14628-14633; Beerli et al., (2000 Proc Natl Acad Sci USA 97 (4), 1495-1500; Beerli et al., (2000) J. Biol. Chem. 275 (42), 32617-32627]; VP64: VP16 minimal activation domain of herpes simplex virus And the fusion domain to the activation domain VP64 or repression domain KRAB of Kox-1 and the pcDNA3 or retroviral pMX-IRES-GFP vector ([Liu et al., (1997) Proc Natl Acad Sci USA 94, 10669-10673]; IRES, internal ribosome-entry site; GFP, green fluorescent protein).

Example 3: General Methods Transfection and Luciferase Assay HeLa cells were used at 40-60% of culture saturation density. Cells were transfected with 160 ng reporter plasmid (pGL3-promoter construct) and 40 ng effector plasmid (zinc finger-effector domain fusion in pcDNA3) in 24-well plates. Cell extracts were prepared 48 hours after transfection and measured with a luciferase assay reagent (Promega) in a MicroLumat LB96P luminometer (EG & Barsold, Geysersburg, MD).

Retroviral gene targeting and flow cytometric analysis Reported procedures [Beerli et al., (2000) Proc Natl Acad Sci USA 97 (4), 1495-1500; Beerli et al., (2000) J. Biol. Chem. 275 (42), 32617-32627] performed these assays. As primary antibodies, ErbB-1 specific mAb EGFR (Santa Cruz), ErbB-2 specific mAb FSP77 (contributed by Nancy E. Hynes; Harwewrth et al., 1992) and ErbB-3 specific mAb SGP1 (Oncogene Research)・ Products were used. Fluorescently labeled donkey F (ab ′) 2 anti-mouse IgG was used as a secondary antibody (Jackson Immuno-Research).

Computer modeling A computer model was created using Insight II (Molecular Simulations, Inc.). Model was based on Zif268-DNA (PDB accession 1AAY) and QGSR- GCA C coordinates of the co-crystal structure of (SEQ ID NO: 107) (1A1H). These structures are not energy minimized and only suggest possible interactions if presented. Hydrogen bonds were considered to be reliable when the distance between heavy atoms was 3 (+/− 0.3) angstroms and the angle formed by heavy atoms and hydrogen was 1200 or greater. The van der Waals interaction, which appears to be reliable, required a distance between 4 (+/− 0.3) angstrom methyl group carbon atoms.

Claims (17)

A synthetic polypeptide comprising _{2 to} 12 Cys ₂ -His ₂ zinc finger-nucleotide binding peptides, wherein the synthetic polypeptide has the sequence 5 ′-(ANN) _n −3 ′, wherein each N is A, C, G or T, and n is 2 to 6), wherein at least one of the binding peptides is any sequence of SEQ ID NOs: 7-70 and 109 A synthetic polypeptide comprising. The synthetic polypeptide of claim 1, wherein the at least one nucleotide binding peptide has the sequence of any of SEQ ID NOs: 7-70 and 109. 2-6 Cys ₂ -His ₂ zinc finger - containing nucleotide binding peptide, a synthetic polypeptide of claim 1, wherein. The synthetic polypeptide of claim 1, wherein each of the nucleotide binding peptides binds to a different DNA sequence. That the peptide binds to a selected DNA sequence comprising the sequence 5 ′-(ANN) _n −3 ′, wherein each N is A, C, G or T and n is 2-6. 4. A synthetic polypeptide according to claim 3, which is capable.   The synthetic polypeptide of claim 1, further operably linked to one or more transcriptional regulators. The synthetic polypeptide of claim 1, wherein each of the nucleotide binding peptides comprises a sequence of any of SEQ ID NOs: 46-70. The synthetic polypeptide of claim 1, wherein each of the nucleotide binding peptides comprises a sequence of any of SEQ ID NOs: 7-45. The synthetic polypeptide of claim 1, wherein each of the nucleotide binding peptides comprises a sequence of any of SEQ ID NOs: 10, 11, 17, 19, 21, 23-30, 32, 34-36, 42, 43 or 45. . 2. The poly of claim 1, wherein each nucleotide binding peptide comprises the sequence of any of SEQ ID NOs: 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58 or 59. peptide.   An isolated and purified polynucleotide encoding the polypeptide of claim 1.   An expression vector comprising the polynucleotide according to claim 11. A method for regulating expression of a selected DNA sequence comprising the sequence (5'-ANN) _n- 3 ', wherein n is an integer from 2 to 12, comprising the effective amount of the polypeptide of claim 1 Exposing the DNA sequence to. 14. The method according to claim 13, wherein the sequence 5 '-(ANN) _n- 3' is present in the transcription region of the DNA sequence. 14. A method according to claim 13, wherein the sequence 5 '-(ANN) _n- 3' is present in the promoter region of the DNA sequence. 14. The method of claim 13, wherein the sequence 5 '-(ANN) _n- 3' is present in the expressed sequence tag.   14. The method of claim 13, wherein the polypeptide is operably linked to one or more transcriptional regulators.

Images